
- •QoS Overview
- •“Do I Know This Already?” Quiz
- •QoS: Tuning Bandwidth, Delay, Jitter, and Loss Questions
- •Foundation Topics
- •QoS: Tuning Bandwidth, Delay, Jitter, and Loss
- •Bandwidth
- •The clock rate Command Versus the bandwidth Command
- •QoS Tools That Affect Bandwidth
- •Delay
- •Serialization Delay
- •Propagation Delay
- •Queuing Delay
- •Forwarding Delay
- •Shaping Delay
- •Network Delay
- •Delay Summary
- •QoS Tools That Affect Delay
- •Jitter
- •QoS Tools That Affect Jitter
- •Loss
- •QoS Tools That Affect Loss
- •Summary: QoS Characteristics: Bandwidth, Delay, Jitter, and Loss
- •Voice Basics
- •Voice Bandwidth Considerations
- •Voice Delay Considerations
- •Voice Jitter Considerations
- •Voice Loss Considerations
- •Video Basics
- •Video Bandwidth Considerations
- •Video Delay Considerations
- •Video Jitter Considerations
- •Video Loss Considerations
- •Comparing Voice and Video: Summary
- •IP Data Basics
- •Data Bandwidth Considerations
- •Data Delay Considerations
- •Data Jitter Considerations
- •Data Loss Considerations
- •Comparing Voice, Video, and Data: Summary
- •Foundation Summary
- •QoS Tools and Architectures
- •“Do I Know This Already?” Quiz
- •QoS Tools Questions
- •Differentiated Services Questions
- •Integrated Services Questions
- •Foundation Topics
- •Introduction to IOS QoS Tools
- •Queuing
- •Queuing Tools
- •Shaping and Policing
- •Shaping and Policing Tools
- •Congestion Avoidance
- •Congestion-Avoidance Tools
- •Call Admission Control and RSVP
- •CAC Tools
- •Management Tools
- •Summary
- •The Good-Old Common Sense QoS Model
- •GOCS Flow-Based QoS
- •GOCS Class-Based QoS
- •The Differentiated Services QoS Model
- •DiffServ Per-Hop Behaviors
- •The Class Selector PHB and DSCP Values
- •The Assured Forwarding PHB and DSCP Values
- •The Expedited Forwarding PHB and DSCP Values
- •The Integrated Services QoS Model
- •Foundation Summary
- •“Do I Know This Already?” Quiz Questions
- •CAR, PBR, and CB Marking Questions
- •Foundation Topics
- •Marking
- •IP Header QoS Fields: Precedence and DSCP
- •LAN Class of Service (CoS)
- •Other Marking Fields
- •Summary of Marking Fields
- •Class-Based Marking (CB Marking)
- •Network-Based Application Recognition (NBAR)
- •CB Marking show Commands
- •CB Marking Summary
- •Committed Access Rate (CAR)
- •CAR Marking Summary
- •Policy-Based Routing (PBR)
- •PBR Marking Summary
- •VoIP Dial Peer
- •VoIP Dial-Peer Summary
- •Foundation Summary
- •Congestion Management
- •“Do I Know This Already?” Quiz
- •Queuing Concepts Questions
- •WFQ and IP RTP Priority Questions
- •CBWFQ and LLQ Questions
- •Comparing Queuing Options Questions
- •Foundation Topics
- •Queuing Concepts
- •Output Queues, TX Rings, and TX Queues
- •Queuing on Interfaces Versus Subinterfaces and Virtual Circuits (VCs)
- •Summary of Queuing Concepts
- •Queuing Tools
- •FIFO Queuing
- •Priority Queuing
- •Custom Queuing
- •Weighted Fair Queuing (WFQ)
- •WFQ Scheduler: The Net Effect
- •WFQ Scheduling: The Process
- •WFQ Drop Policy, Number of Queues, and Queue Lengths
- •WFQ Summary
- •Class-Based WFQ (CBWFQ)
- •CBWFQ Summary
- •Low Latency Queuing (LLQ)
- •LLQ with More Than One Priority Queue
- •IP RTP Priority
- •Summary of Queuing Tool Features
- •Foundation Summary
- •Conceptual Questions
- •Priority Queuing and Custom Queuing
- •CBWFQ, LLQ, IP RTP Priority
- •Comparing Queuing Tool Options
- •“Do I Know This Already?” Quiz
- •Shaping and Policing Concepts Questions
- •Policing with CAR and CB Policer Questions
- •Shaping with FRTS, GTS, DTS, and CB Shaping
- •Foundation Topics
- •When and Where to Use Shaping and Policing
- •How Shaping Works
- •Where to Shape: Interfaces, Subinterfaces, and VCs
- •How Policing Works
- •CAR Internals
- •CB Policing Internals
- •Policing, but Not Discarding
- •Foundation Summary
- •Shaping and Policing Concepts
- •“Do I Know This Already?” Quiz
- •Congestion-Avoidance Concepts and RED Questions
- •WRED Questions
- •FRED Questions
- •Foundation Topics
- •TCP and UDP Reactions to Packet Loss
- •Tail Drop, Global Synchronization, and TCP Starvation
- •Random Early Detection (RED)
- •Weighted RED (WRED)
- •How WRED Weights Packets
- •WRED and Queuing
- •WRED Summary
- •Flow-Based WRED (FRED)
- •Foundation Summary
- •Congestion-Avoidance Concepts and Random Early Detection (RED)
- •Weighted RED (WRED)
- •Flow-Based WRED (FRED)
- •“Do I Know This Already?” Quiz
- •Compression Questions
- •Link Fragmentation and Interleave Questions
- •Foundation Topics
- •Payload and Header Compression
- •Payload Compression
- •Header Compression
- •Link Fragmentation and Interleaving
- •Multilink PPP LFI
- •Maximum Serialization Delay and Optimum Fragment Sizes
- •Frame Relay LFI Using FRF.12
- •Choosing Fragment Sizes for Frame Relay
- •Fragmentation with More Than One VC on a Single Access Link
- •FRF.11-C and FRF.12 Comparison
- •Foundation Summary
- •Compression Tools
- •LFI Tools
- •“Do I Know This Already?” Quiz
- •Foundation Topics
- •Call Admission Control Overview
- •Call Rerouting Alternatives
- •Bandwidth Engineering
- •CAC Mechanisms
- •CAC Mechanism Evaluation Criteria
- •Local Voice CAC
- •Physical DS0 Limitation
- •Max-Connections
- •Voice over Frame Relay—Voice Bandwidth
- •Trunk Conditioning
- •Local Voice Busyout
- •Measurement-Based Voice CAC
- •Service Assurance Agents
- •SAA Probes Versus Pings
- •SAA Service
- •Calculated Planning Impairment Factor
- •Advanced Voice Busyout
- •PSTN Fallback
- •SAA Probes Used for PSTN Fallback
- •IP Destination Caching
- •SAA Probe Format
- •PSTN Fallback Scalability
- •PSTN Fallback Summary
- •Resource-Based CAC
- •Resource Availability Indication
- •Gateway Calculation of Resources
- •RAI in Service Provider Networks
- •RAI in Enterprise Networks
- •RAI Operation
- •RAI Platform Support
- •Cisco CallManager Resource-Based CAC
- •Location-Based CAC Operation
- •Locations and Regions
- •Calculation of Resources
- •Automatic Alternate Routing
- •Location-Based CAC Summary
- •Gatekeeper Zone Bandwidth
- •Gatekeeper Zone Bandwidth Operation
- •Single-Zone Topology
- •Multizone Topology
- •Zone-per-Gateway Design
- •Gatekeeper in CallManager Networks
- •Zone Bandwidth Calculation
- •Gatekeeper Zone Bandwidth Summary
- •Integrated Services / Resource Reservation Protocol
- •RSVP Levels of Service
- •RSVP Operation
- •RSVP/H.323 Synchronization
- •Bandwidth per Codec
- •Subnet Bandwidth Management
- •Monitoring and Troubleshooting RSVP
- •RSVP CAC Summary
- •Foundation Summary
- •Call Admission Control Concepts
- •Local-Based CAC
- •Measurement-Based CAC
- •Resources-Based CAC
- •“Do I Know This Already?” Quiz
- •QoS Management Tools Questions
- •QoS Design Questions
- •Foundation Topics
- •QoS Management Tools
- •QoS Device Manager
- •QoS Policy Manager
- •Service Assurance Agent
- •Internetwork Performance Monitor
- •Service Management Solution
- •QoS Management Tool Summary
- •QoS Design for the Cisco QoS Exams
- •Four-Step QoS Design Process
- •Step 1: Determine Customer Priorities/QoS Policy
- •Step 2: Characterize the Network
- •Step 3: Implement the Policy
- •Step 4: Monitor the Network
- •QoS Design Guidelines for Voice and Video
- •Voice and Video: Bandwidth, Delay, Jitter, and Loss Requirements
- •Voice and Video QoS Design Recommendations
- •Foundation Summary
- •QoS Management
- •QoS Design
- •“Do I Know This Already?” Quiz
- •Foundation Topics
- •The Need for QoS on the LAN
- •Layer 2 Queues
- •Drop Thresholds
- •Trust Boundries
- •Cisco Catalyst Switch QoS Features
- •Catalyst 6500 QoS Features
- •Supervisor and Switching Engine
- •Policy Feature Card
- •Ethernet Interfaces
- •QoS Flow on the Catalyst 6500
- •Ingress Queue Scheduling
- •Layer 2 Switching Engine QoS Frame Flow
- •Layer 3 Switching Engine QoS Packet Flow
- •Egress Queue Scheduling
- •Catalyst 6500 QoS Summary
- •Cisco Catalyst 4500/4000 QoS Features
- •Supervisor Engine I and II
- •Supervisor Engine III and IV
- •Cisco Catalyst 3550 QoS Features
- •Cisco Catalyst 3524 QoS Features
- •CoS-to-Egress Queue Mapping for the Catalyst OS Switch
- •Layer-2-to-Layer 3 Mapping
- •Connecting a Catalyst OS Switch to WAN Segments
- •Displaying QoS Settings for the Catalyst OS Switch
- •Enabling QoS for the Catalyst IOS Switch
- •Enabling Priority Queuing for the Catalyst IOS Switch
- •CoS-to-Egress Queue Mapping for the Catalyst IOS Switch
- •Layer 2-to-Layer 3 Mapping
- •Connecting a Catalyst IOS Switch to Distribution Switches or WAN Segments
- •Displaying QoS Settings for the Catalyst IOS Switch
- •Foundation Summary
- •LAN QoS Concepts
- •Catalyst 6500 Series of Switches
- •Catalyst 4500/4000 Series of Switches
- •Catalyst 3550/3524 Series of Switches
- •QoS: Tuning Bandwidth, Delay, Jitter, and Loss
- •QoS Tools
- •Differentiated Services
- •Integrated Services
- •CAR, PBR, and CB Marking
- •Queuing Concepts
- •WFQ and IP RTP Priority
- •CBWFQ and LLQ
- •Comparing Queuing Options
- •Conceptual Questions
- •Priority Queuing and Custom Queuing
- •CBWFQ, LLQ, IP RTP Priority
- •Comparing Queuing Tool Options
- •Shaping and Policing Concepts
- •Policing with CAR and CB Policer
- •Shaping with FRTS, GTS, DTS, and CB Shaping
- •Shaping and Policing Concepts
- •Congestion-Avoidance Concepts and RED
- •WRED
- •FRED
- •Congestion-Avoidance Concepts and Random Early Detection (RED)
- •Weighted RED (WRED)
- •Flow-Based WRED (FRED)
- •Compression
- •Link Fragmentation and Interleave
- •Compression Tools
- •LFI Tools
- •Call Admission Control Concepts
- •Local-Based CAC
- •Measurement-Based CAC
- •Resources-Based CAC
- •QoS Management Tools
- •QoS Design
- •QoS Management
- •QoS Design
- •LAN QoS Concepts
- •Catalyst 6500 Series of Switches
- •Catalyst 4500/4000 Series of Switches
- •Catalyst 3550/3524 Series of Switches
- •Foundation Topics
- •QPPB Route Marking: Step 1
- •QPPB Per-Packet Marking: Step 2
- •QPPB: The Hidden Details
- •QPPB Summary
- •Flow-Based dWFQ
- •ToS-Based dWFQ
- •Distributed QoS Group–Based WFQ
- •Summary: dWFQ Options

Link Fragmentation and Interleaving 499
Now imagine that you need to configure R1 in the figure to use MLP LFI. You already know that you want a maximum serialization delay of 10 ms, and conveniently, MLP LFI enables you to configure a max-delay parameter. MLP LFI then calculates the fragment size, based on the following formula:
Max-delay * bandwidth
In this formula, bandwidth is the value configured on the bandwidth interface subcommand, and max-delay is the serialization delay configured on the ppp multilink fragment-delay command. For instance, R1 in Figure 7-8 shows a budget for 10 ms of serialization delay. On a 56-kbps link, a 10-ms max-delay would make the fragment size 56,000 * .01, or 560 bits, which is 70 bytes.
Cisco generally suggests a maximum serialization delay per link of 10 ms in multiservice networks. Because serialization delay becomes less than 10 ms for 1500-byte packets at link speeds greater than 768 kbps, Cisco recommends that LFI be considered on links with a 768-kbps clock rate and below.
The math used to find the fragment size, based on the serialization delay and bandwidth, is pretty easy. For perspective, Table 7-8 summarizes the calculated fragment sizes based on the bandwidth and maximum delay.
Table 7-8 Fragment Sizes Based on Bandwidth and Serialization Delay
Bandwidth/Link Speed |
10-ms Delay |
20-ms Delay |
30-ms Delay |
40-ms Delay |
|
|
|
|
|
56 kbps |
70 |
140 |
210 |
280 |
|
|
|
|
|
64 kbps |
80 |
160 |
240 |
320 |
|
|
|
|
|
128 kbps |
160 |
320 |
480 |
560 |
|
|
|
|
|
256 kbps |
320 |
640 |
960 |
1280 |
|
|
|
|
|
512 kbps |
640 |
1280 |
1920* |
2560* |
|
|
|
|
|
768 kbps |
1000 |
2000* |
3000* |
4000* |
|
|
|
|
|
1536 kbps |
1600* |
3200* |
4800* |
6400* |
|
|
|
|
|
*Values over 1500 exceed the typical maximum transmit unit (MTU) size of an interface. Fragmentation of sizes larger than MTU does not result in any fragmentation.
Frame Relay LFI Using FRF.12
Cisco IOS Software supports two flavors of Frame Relay LFI. The more popular option, FRF.12, is based on Frame Relay Forum Implementation Agreement 12, with the other option, FRF.11-C, being based on Frame Relay Forum Implementation Agreement 11, Annex C. FRF.12 applies to data VCs, and FRF.11-C applies to voice VCs. Because most Frame Relay VCs are data VCs, and because most service providers do not offer FRF.11 (VoFR) VCs, this section focuses on FRF.12. The final part of the FRF configuration section at the end of this chapter covers the differences between FRF.12 and FRF.11-C.

500 Chapter 7: Link-Efficiency Tools
NOTE |
Another LFI feature, called multilink PPP over Frame Relay and ATM, also provides an option |
|
for LFI. This option is suited for environments that use Frame Relay/ATM internetworking and |
|
desire to run delay-sensitive applications such as VoIP on slow-speed WAN links. |
|
|
FRF.12 varies greatly from MLP LFI in terms of how it works with queuing tools. IOS requires that Frame Relay traffic shaping (FRTS) be used to also use FRF.12. Remember all the trivia from Chapter 5, “Traffic Policing and Shaping”? You may recall that FRTS can apply queuing tools to shaping queues associated with each VC, but FRTS only allows a single FIFO queue on the physical interface. When you add FRF.12 to FRTS, however, two interface FIFO output queues are created rather than the single FIFO queue. Figure 7-9 shows the two FIFO interface output queues, called Dual FIFO queues, with FTRS and FRF.12.
Figure 7-9 Interface Dual FIFO Queues with FRTS and FRF.12
1500 Byte Packet Arrives, Followed by One 60 Byte Packet
|
|
“Dual FIFO” Queue 1 |
|
|
|
|
|
|
|
Small |
|
TX Queue, Length 2 |
|
|
|
|
Packet |
|
||
|
|
|
|
|
|
|
Fragment |
|
|
|
Schedule |
Frag 2 |
Frag1 |
Classify |
|
|
Like PQ |
|||
|
|
|
|
|||
if > 300 |
|
|
|
|
|
|
Frag 5 |
Frag 4 |
Frag 3 |
|
|
|
|
|
|
|
|
|||
|
|
“Dual FIFO” Queue 2 |
|
|
|
Figure 7-9 focuses on the interface output queues, ignoring the shaping queues. Just like in the figure that depicted MLP LFI, a 1500-byte packet arrives, followed by a 60-byte packet. The large packet is fragmented into five 300-byte packets, with the first two being placed into the TX Queue, and the last three ending up in one of the interface output queues. The small packet arrives next, and it is not fragmented, because it is less than 300 bytes in length. It is placed into the other Dual FIFO queue.
This two-queue Dual FIFO structure acts like the queuing tools described in Chapter 4 in many ways. It has classification logic that places packets into one queue or the other (more on that in a few paragraphs). It has a number of queues (always two), and it has particular behavior inside each queue (FIFO). It also performs scheduling between the two queues using an algorithm

Link Fragmentation and Interleaving 501
such as Priority Queuing’s (PQ) scheduling algorithm. Therefore, to understand what happens, you need to take a closer look at the classification logic and the scheduling algorithm applied to the Dual FIFO queues.
First, when a packet passes through the fragmentation step in Figure 7-9, if there are no packets in either Dual FIFO queue, and there is room in the TX Queue/TX Ring, the fragments get placed into the TX Ring/TX Queue until it is full. That’s why in Figure 7-9 the first two fragments of the large packet got placed into the 2-entry TX Queue. Then, when the TX Queue is full, packets are placed into one of the two Dual FIFO queues.
IOS schedules packets from the Dual FIFO interface queues into the interface TX Queue in a PQ-like fashion. The logic treats one of the two Dual FIFO queues like the PQ High queue, and the other like the PQ Normal queue. The scheduler always takes packets from the High queue first if one is available; otherwise, the scheduler takes a packet from the Normal queue. Just like PQ, the scheduler always checks the High queue for a packet before checking the Normal queue. Although IOS does not give a lot of information about the two Dual FIFO queues in show commands, one command (show queueing interface) does list counters for the High and Normal queues. (This book refers to these two queues as the High and Normal Dual FIFO queues, even though most other IOS documents and courses do not even name the two queues.)
The classification logic with the FRF.12 Dual FIFO queues appears to be very straightforward. One popular school of thought concerns how FRF.12 classifies packets into the Dual FIFO queues, however, and another less-popular school of thought describes how classification really works. Most of the available references, including the courses on which the exams are based, state that the classification boils down to this:
•
•
Fragmented packets are placed in the Normal Dual FIFO queue.
Unfragmented packets are placed in the High Dual FIFO queue.
Putting the classification logic together with the queue service logic makes one neat package. LFI wants to interleave the small packets between fragments of the larger packets. By classifying the unfragmented packets into the High queue, and the fragments into the Normal queue, the PQ-like queue service algorithm interleaves unfragmented packets in front of fragmented packets.
In spite of what the courses actually say, FRF.12 actually classifies packets into one of the Dual FIFO interface output queues based on the queuing configuration for the shaping queues on each VC. FRTS allows a large variety of queuing tools to be configured for the shaping queues. Two of these queuing tools, if enabled on the shaping Queue of a VC, cause packets to be placed in the High Dual FIFO queue on the physical interface. Figure 7-10 outlines the main concept.

502 Chapter 7: Link-Efficiency Tools
Figure 7-10 Classification Between FRTS LLQ Shaping Queues and Interface Dual FIFO Queues with FRF.12
Shaping Queues Created |
|
|
|
by LLQ Configuration |
|
|
|
on a Single VC |
|
FRF.12 Dual FIFO Queues |
|
|
|
||
Class 1 – |
|
|
|
LLQ |
|
|
|
|
|
High Queue |
|
Class 2 – |
|
TX Ring |
|
Non-LLQ |
|
||
Shape |
Normal |
||
|
|||
|
|
||
|
|
Queue |
|
Class 3 – |
|
|
|
Non-LLQ |
|
|
|
Class 3 – |
|
|
|
Non-LLQ |
|
|
The figure depicts LLQ for the shaping queue on a single VC feeding into the interface Dual FIFO queues. The shaping logic remains unchanged, as does the LLQ logic for the shaping queues—in other words, with or without FRF.12 configured, the behavior of shaping acts the same. The only difference created by adding FRF.12 to FRTS comes when FRTS must decide which of the two interface Dual FIFO queues to place the packet into after the shaper allows it to pass. (Without FRF.12, a single FIFO interface queue exists, in which case classification logic is not needed.)
As shown in the figure, the only way a packet makes it to the High Dual FIFO queue is to have first been in the low-latency queue. In other words, FRF.12 determines which packets are interleaved based on which packets were placed into the low-latency queue in the shaping queue.
The classification logic and the scheduling logic make perfect sense if you consider the packets that need the minimal latency. When you purposefully configure LLQ for shaping queues, the class of packets you place into the low-latency queue must be the ones for which you want to minimize latency. FRF.12 should interleave those same packets to further reduce latency; therefore, FRF.12 just places those same packets into the Dual FIFO High queue.
NOTE Because the Dual FIFO queues created by FRF.12 essentially creates a high-priority queue appropriate for VoIP traffic, when you are using FRTS, Cisco also recommends configuring LFI on links that run at speeds greater than 768 kbps. However, you should configure the fragment size to something larger than the MTU—for instance, 1500 bytes. By doing so, no packets are actually fragmented, but VoIP packets can be placed in the high-priority queue in the Dual FIFO queuing system on the physical interface.